Biological Control of a Tropical Weed: A Population Model and Experiment for Sida acuta

Abstract

1. The effects of defoliation of the malvaceous tropical weed Sida acuta by an introduced biological control agent, the chrysomelid Calligrapha pantherina, were investigated in northern Australia with an insecticidal exclusion experiment. 2. Calligrapha pantherina was found to reduce annual seed production by an order of magnitude, from 8001 seeds m -2 to 731m -2 . However, there was no measurable effect on individual survival, mass per seed, or total biomass of the weed in the year of defoliation. 3. A model for annual plants was adapted and used to predict the density at flowering of the weed in the following year that would result from the measured reduction in seed output. The model predicted a fall in density from 76.0 plants m -2 to 42.0 m -2 in the year following defoliation. The actual density was 50.2 m -2 , not significantly different from the predicted value. 4. If this level of reduction in seed output were maintained, the density of the weed would continue to fall to a level that would depend on the searching efficiency of the beetle. However, there are still considerable uncertainties in the estimation of some of the model's plant parameters, particularly for seed losses. Moreover, the grazing and population dynamics of the beetle remain to be determined. 5. The model is used to show that the presence of seed carry-over (or generational dormancy) usually lowers plant density but, where herbivory is patchy, it reduces the rate of extinction of populations.